CABLES rAdiAting Application note
CABL
ESrAdiAting CABL
ES
Application note
➔ INTRODUCTION 3
➔ 1.LONGITUDINALATTENUATIONANDCOUPLINGLOSS 3
➔ 2.RADIATEDANDCOUPLEDMODECABLES 5
➔ 3.LINKBUDGET 73.1.RCinsertionloss 83.2.RCCouplingloss 83.3.Correctionforlongerdistance 103.4.Penetrationloss 113.5.Mobileantennalossrelativetodipole 113.6.Safetymargin 11
➔ 4.COUPLINGLOSS 124.1.Measurementprocedures 12
4.1.1.Ground–levelversusfree-spacemethod 124.1.2.Ground–levelversusor-spacemethod 154.1.3.EupenRMCRange 15
4.2Couplinglossandantennaorientations 164.2.1.CLdefinition accordingtothestandard 164.2.2.Antennaorientationandlinkbudget 18
➔ 5.RCPERFORMANCESOPTIMISATION 195.1.RCpositioning 19
5.1.1.Mobileantennamountedonthevehicleroof 195.1.2.Hand-heldmobileequipmentonboardtrain 20
5.2.Multi-cablesystem 215.3.Resonantfrequencies 22
Listofabbreviations
CL designatesacouplinglossingeneral,whichevertheorientationandtheprobabilitylevel(50%,95%oranyotherpercentile).
CL50% medianvalueofthecouplingloss.
CL95% 95%percentileofthecouplingloss.
CLr,CLpandCLo Inthecontextof§3.2and4.2only,thesesymbolsdesignatetheCLsintheradial,parallelandorthogonalorientationsrespectively,whichevertheprobabilitylevel.Theseabbreviationshavenotbeenusedeverywheretosimplifythemathematicalformulas.
CLmean Inthecontextof§3.2and4.2only,thissymbolcorrespondstothemeanCLaveragedoverthethreeantennaorientationsasdefine inthestandard.
RC radiatingcable.
RadiatingCablesApplicationnote
11/2021
TABLEOFCONTENTS
3
INTRODUCTION
RadiatingCablesApplicationnote11/2021
Theaimofthisapplicationnoteistoprovideusefulinformationfor:
➔ theperformancesoptimisationoftheEupenradiatedmodecables;➔ reliablelinkbudget;➔ RCperformancescomparison.
Section1and2 includeabriefreminderofsome importantdefinitions Howtoperforma linkbudgetcalculationisexplainedinsection3.
The radio engineers familiar with the RC subject have certainly noticed that there is a real lack ofharmonisationconcerningthedefinitio ofthecouplingloss.Indeed,documentssuchasdatasheetsandapplicationnotespublishedbytheRCmanufacturersrevealdifferencesofinterpretationthatmayleadtosignifican errorsinlinkbudgetorwhentheRCperformanceshavetobecompared.ThesedifferencesarerathersurprisingasalltheRCmanufacturersrefertothesameIECstandard.
Thevariousmethodstomeasurethecoupling lossarepresented insection4. Itssensitivitytoantennaorientationandotherparametersisdeeplyanalysed.Suchaninformationmaybeusefulforlinkbudgetcalculations.
SomerulesallowingoptimisationofRCperformancesarealsopresentedinsection5.
1. LONGITUDINALATTENUATIONANDCOUPLINGLOSS
Fromtheelectricalpointofview,RCperformancesaremostlycharacterisedbythelongitudinalattenuation(indB/100m)andbythecouplingloss(indB).
Thelongitudinal attenuationisameasureoftheattenuationofthesignalpropagatinginsidetheRC.Itisspecifie indBperunitlength(usuallyindB/100m)andisgivenbythefollowingformula:
wherePinandPoutaretheRCinputandoutputpowersrespectively.
Thelongitudinalattenuationisprimarilytheresultscopperanddielectriclossesandamountofradiatedenergy.Thelongitudinalattenuationincreaseswiththefrequencyanddecreaseswiththecablediameter.Itisalsosomewhatinfluence bytheproximityoftheRCtoothersurfaces.
Thecoupling loss (CL)characterises thecouplingbetween theenergy travelling inside theRCandareceivingantenna. It isdefine astheratioofthereceivedpowerattheantennaoutputtothepowerfl wingintheRC.Forexample,ifthepowerfl wingintheRCwas0dBmandthepowerreceivedbytheantennawas–60dBm,thentheCLwouldbe60dB.Inthedatasheet,theCLisgivenforanRCtoantennadistanceequalto2m.
Pin
➔ a=10log——Pout
4
ThelocalvalueoftheCLisgivenbythefollowingformula:
wherePcableisthepowerinsidetheRC(neartheantenna)andPantennathepowerattheantennaoutput.
Usually,CL50%andCL95%arespecifie intheRCdatasheets.TheirmeaningisillustratedinFigure1.Thecurverepresentstheprofil ofthesignal(indBm)receivedbyanantennamovedalongapathparalleltotheRC,forexampleat2m.Thehorizontallineatthetopofthediagramrepresentsthepower(indBm)insidetheRC.Thedistance(indB)betweenthehorizontallineandthecurveisequaltotheCLatthisparticularpoint.TheCL50%correspondstothe50%percentileormedianvalue.Itmeansthat50%ofthemeasuredlocalvaluesarelowerand50%arehigher.
TheCL95%correspondstothe95%percentile.Itmeansthat95%ofthemeasuredlocalvaluesarelowerthanthisfigu e.
TheCLmeasurementmethodsaccordingtotheIECstandardaredetailedandanalysedinsection4.
Figure1:CL50%andCL95%definition
Pcable
➔ CL=10log————Pantenna
CL95%
CL50%
powerinsidethecablepowerreceivedbytheantenna
RadiatingCablesApplicationnote
11/2021
5
2. RADIATEDANDCOUPLEDMODECABLES
AllofthedifferenttypesofRCsarebasedontheeffectsofsphericalwavesexcitedbytheleakageproducedbytheaperturesintheexternalconductor.Theresultingfiel atacertaindistanceisgivenbythevectoradditionalloftheaperturescontributions.TheFigure2showsasimplecasewheretheelectromagneticwavesgeneratedbyonlythreeaperturesareconsidered.Thesewavesareidentifie bythesymbolW1,W2andW3.Let’sconsidertheresultingelectromagneticfiel atapointP.
ThevectorsE1,E2andE3inFigure3representtheelectricfiel componentatthispointPcorrespondingtoW1,W2andW3respectively.ThevectorEistheirresultant.Withoutspecialprecautions,theaperturecontributionsmaygiverisetodestructiveinterferencesatsomeplaceswithalowresultingfiel indicatedbyarelativelyshortvectorasshowninFigure3(leftside).Conversely,aperturecontributionsmaybeinphaseatotherplaces,whichgiverisetoconstructive interferences,henceastrongresultantasshowninFigure3(rightside).ItisclearthattherationaleinFigure3appliestobothelectricandmagneticfiel components.
Therealsituation isobviouslymorecomplexastheelectromagneticfiel atanypoint istheresultsofmorethan3aperturescontributions.Therationaleishoweveridenticalanditexplainsthefluctuation ofthefiel strengthalongtheRC.Typically,thesefluctuation reach20to30dBpeaktopeakandcanbemodelisedbyaRayleighdistribution.Usually,thedifferencebetweenCL95%andCL50%rangesbetween10and13dB.
Figure2:Electromagneticwaves
duetotheaperturesintheexternalconductor
W1W2W3
P
RadiatingCablesApplicationnote11/2021
6
Theradiated modeRCsaredesignedtoproduceacoherent interferenceofthedifferentaperturescontributions incertainfrequencybandsandatallplacesaroundtheRC.Thiseffect isobtained iftheaperturespacingischoseninsuchawaythatalltheaperturecontributionsaddupinphaseintheRCradiocoverageareaasshowninFigure3(rightside).Thisisachievedifthedelaybetweenthecontributionsoftwosuccessiveaperturesisamultipleofthesignalperiod.Whenthisconditionissatisfied theresultantfiel isstrongerandthefiel strengthfluctuation alongtheRClengthareconsiderablyreduced.Itresultsthat:
➔ theCL50%decreases;➔ thedifferencebetweenCL95%andCL50%decreasesandtypicallyrangesfrom3to8dB.
Withradiatedmodecables,themaindifficult istomaximisethefrequencybandinwhichtheaperturescontributionsinterfereinacoherentway.
AlltheradiatedmodeRCsworkincoupledmodebelowacertainfrequency,hereaftertermed“transition frequency”.Thistransitionislinkedtotheaperturespacing.Thisisbecauseitisimpossibletokeepthedifferentaperturescontributionsinphasewhenthewavelengthexceeds,approximately,twotimesthedistance between two successive aperture groups. However, the performances may be impaired, forvariousreasons,insomefrequencybandsabovethetransitionfrequency.Thismeansthattheradiatedmodeisnotnecessarily“superior”thancoupledmode.
TheEupenradiatedmodeRCsaredesignedtoprovidelowCLandlowfiel strengthfluctuation inseveralfrequencybandsallocatedtomostmobileradiosystemsstandardssuchasTETRA,TETRAPOL,GSM900,GSMR,GSM1800,PCS,DECT,UMTS,WiFi(eitherat2.4GHzorat5to6GHz),WiMax,etc.
E1
^
E2
^
E
^E
^
E2
^
E1
^
E3
^E3
^
Ě1,Ě2andĚ3notinphase Ě1,Ě2andĚ3nearlyinphaseFigure3:Aperturecontributionsoutofphase(left)andinphase(right)
RadiatingCablesApplicationnote
11/2021
7
3. LINKBUDGET
The basic elements to calculate a link budget can be illustrated by considering the example showninFigure4. It involvesaGSM900radiocoverage inadual-boretunnelthat is900m in length. Itshallbeassumedthat:
➔ thepowerperchannelavailableforthedownlinkis1W(+30dBm);➔ theRC ineachbore is fedviaapower splitter (“T-feed”configu ation) the insertion lossofwhich
isequalto3.5dB;➔ jumper cables are used to connect the repeater, power splitter and the RCs. Their total insertion
lossisequalto1.5dB;➔ thespecificatio imposesa receivedsignal (measuredwithahalf-wavedipoleantenna)ofat least
-88dBmat95%ofthepointsinthevicinityofthecableendandat6mdistance.
It shall also be assumed that the RC has the following characteristics at 960 MHz (upper limit of theGSM900frequencyband):
➔ longitudinalattenuation:3.1dB/100m;➔ CL50%andCL95%equalto58and62dBrespectively.
Table1summarisesthelinkbudget.Thelastlineindicatesthatthespecificatio issatisfied i.e.aminimumreceivedsignal(measuredwithahalf-wavedipoleantenna)higherthan-88dBmat95%ofthepointsinthevicinityofthecableendandat6mdistance.Thevariouslinesofthislinkbudgetarecommentedhereafter.
Uplinkperformances(i.e.frommobilestationtobasestation)canbecomputedinthesameway.
Table1:Linkbudgetexample
Availablepowerperchannel + 30dBm
Jumpercableloss - 1.5dB
Powersplitterinsertionloss - 3.5dB
RCinsertionloss:900mwith3.1dB/100m - 28dB
CL95%at2m=62dB - 62dB
Correctionforlongerdistance=20log(d/2)=20log(6/2)= - 9.5dB
Penetrationloss* 0dB
Mobileantennalossrelativetodipole* 0dB
Safetymargin - 10dB
Minimum received signal at 6 m from the RC (95% percentile) - 84.5 dBm
Figure4:Dual-boretunnel
withonebasestationandapowersplitter
*Note:Itresultsfromthespecificatio ofthisparticularexamplethatthepenetrationlossandmobileantennalossrelativetodipoleareequalto0dB.
RCinbore1
RCinbore2
RadiatingCablesApplicationnote11/2021
8
3.1. RC insertion loss
The RC insertion loss is equal to the cable length multiplied by the longitudinal attenuation. Thislongitudinalattenuationissomewhatinfluence bythestandoffdistance(betweentheRCandthewallorceilingtowhich it ishung).Forexample, iftheRC isdirectlyagainstaconcretesurface,the impactonthelongitudinalattenuationisfrequencydependentandisobviouslynotidenticalforallRCs.IftheIECstandardconditionsareconsideredasreferencevalues,measurementscarriedwithvariousEupenRMCsindicatethatinstallingtheRCdirectlyagainstaconcretesurfaceinvolvethefollowinglongitudinalattenuationincreases:
➔ below300MHz,theimpactisnegligibleandevensometimesnegative;➔ typicallyrangesfrom5to10%around450MHz;➔ typicallyrangesfrom10to20%around900MHz;➔ typicallyrangesfrom25to60%around2000MHz.
The longitudinalattenuation isalso influence byhumidityanddustdepositontheRC jacket.Even inrathersevereconditions,thelongitudinalattenuationincreaseneverexceeds10%.
3.2. RC Coupling loss
Some RC manufacturers use the free space method or specify the CLs for the antenna orientationcorrespondingtothebestresult.ThedifferencesofinterpretationinthemeaningoftheCLparametermayleadtosignifican errorsinlinkbudgetorwhentheperformancesofdifferentproductshavetobecompared.
TheCL50%andCL95%specifie intheEupendatasheetsaremeasuredwiththegroundlevelmethodaccordingthe IEC61196-4standard1.Theground levelmethodhasbeenpreferredbecause itdefine conditions which are closer to those actually met in practice. Indeed, in almost all the applications,theRCishungatshortdistancefromasurface(ceilingorwall).Adetailedanalysisofthisissueispresentedinsection4.
However,CLsmeasuredwiththefreespacemethodarealsoavailableforsomeEupenRCs.
TheCL50%andCL95%specifie intheEupendatasheetsareaveragedoverthreeantennaorientations(radial,orthogonalandparallel).Asexplainedinsection4.2.,theCL50%orCL95%thatshouldbeusedforlinkbudgetcorrespondtothesymbolsCL50%-meanorCL95%-mean.
1IEC61196-4standard-Coaxialcommunicationcables-Part4:Sectionalspecificatio forradiatingcables.
RadiatingCablesApplicationnote
11/2021
9
a) E.M.wavedepolarisationduetoreflection onobstacles
In thecaseofcommunicationswithhand-heldmobileequipmentonboard train,awavepenetratingintoacarriageexperiencesreflection onthecarriagewalls,ceiling,floo ,seats,etc.Ateachpoint,thefiel strengthisthevectoradditionofseveralwavesandthepolarisationofthesumcanbeconsideredaselliptical ratherthan linear.Figure5showsthesimplecasewhereadirectwave radiatedbytheRCinterfereswithanotherwavewhichhasbeenreflec edbythecarriageceilingandwindow.Thedashedarrows(atrightanglewiththedirectionofpropagation)indicatethewavepolarisation.Ifweconsiderthedifferenceofpropagationdelay,itisclearthatthepolarisationoftheresultingE.M.fiel atthereceptionpointR isverycomplex.Figure5 isavery simplecasewithonlyone reflec edwave. Inpractice, thesituationsaremuchmorecomplexassuggested inFigure6wherethedirectwavemaybeblockedbytravellersorbyanothertrain.
b) Mobileantennaorientation
Withhand-heldequipments,themobileantennaorientation isneitherperfectlyverticalnorhorizontalbutratheracombinationoftheseasshowninFigure7.Indeed,innormaluse,anhand-heldequipmentisslightlydown-tiltedandnotnecessarilyorientatedformaximumresponse.
Figure5:Wavedepolarisation
duetoreflection
Figure6:Propagationintocarriage
Figure7:Mobileantenna
orientation
Figure6 Figure7
RadiatingCablesApplicationnote11/2021
10
c) Mobilestationareratherinsensitivetoantennaorientation
Asexplainedinsection4.2,thedifferencebetweentheCLsinthedifferentantennaorientationisduetothedirectivityofthehalf-wavedipolewhichisusedforCLmeasurements.
Converselymobilestationantennas(suchasGSM,PCN,UMTS,etc.)aremoresophisticatedthatthesingledipoleormonopole(whichisillustratedinFigures6and7).Theirspatialresponseisdifferentandmuchmore“isotropic”thanquarter-wavemonopoleandhalf-wavedipoleantennas.Measurementsperformedwithvariousmobilestationsdemonstratethatthereceivedpowerisnearlyindependentoftheantennaorientation.So,mobilestationantennasbehaveasnearlyisotropicantennaswhichpickupthestrongestfiel componentwitharatherlowgain(about–10dB).
If the symbols CLr, CLp and CLo designate the coupling losses in the radial, parallel and orthogonalorientationrespectively(whichevertheprobability level)and ifthesymbolCLmeancorrespondstothemeancoupling lossasdefine in the IEC61194-4standard, it isshown insection4.2 thatCLmean isgenerally about 4 dB higher than the lowest coupling loss. It results that, for practice, the followingapproximationcanbemade:
CLmean = min (CLr, CLo, CLp) + 4
wherethe“min”symboldesignatestheminimumofthevaluesinbrackets.
This results means that calculating the link budget with the lowest coupling loss (CLr, CLp or CLo) instead of the average value (CLmean) is equivalent to a 4 dB decrease of the safety margin.
3.3. Correction for longer distance
With“classical”transmittingantennas,thereceivedpowerdecreasesasafunctionofthesquareofthedistanced,i.e.:
This is a consequence of the “spherical symmetry” (the radiated energy is contained in a sphere ofradiusequaltod.WithRCs,theradiatedenergyiscontainedinacylinderofradiusequaltod,hencea“cylindricalsymmetry”.Consequently,thereceivedpowerdecreasesasafunctionofthedistanced,i.e.:
1➔ Prec÷—
d2
1➔ Prec÷—
d
RadiatingCablesApplicationnote
11/2021
11
TheCL50%isspecifie at2moftheRCaccordingtotheIEC61196-4standard.Ifitisrequiredatanotherdistance,thefollowingcorrectionshouldbeapplied:
FortheCL95%,alongerdistanceinvolvesastrongerinfluenc ofscatteredradiationsandreflection onwallsandceiling,henceafadingincrease.Thefollowingcorrectioncanbeapplied:
3.4. Penetration loss
For communications into vehicles, the link budget must take a penetration loss into account. Thispenetrationlossisstronglyinfluence bythefrequency,thewidowsizes,theglasstype(singleordoublelayer) and the possible presence of metal coating (for thermal insulation). For example, at 900 MHz,penetrationlossmayrangefrom2or3dBforasinglelayerglass.Itreaches30dBinthecaseofmetalcoatedglasses.
3.5. Mobile antenna loss relative to dipole
Theantennasusedinmobilephones(suchasGSM,PCN,UMTS,etc.)haveanegativegainwithrespecttothehalfwavedipolenormallyusedtomeasuretheCLs.Theirspatialresponseishowevermore“isotropic”.A10dBmobileantennalossrelativetohalfwavedipoleseemsarealisticvalue.
3.6. Safety margin
A10dBsafetymarginisrecommendedtoaccountfor:
➔ thedifferencesbetweenthestandardconditionsinwhichtheCLsaremeasuredandthoseactuallymetinarealtunnelenvironment;
➔ thevariousfactorswhichmayimpairtheRCperformances.
As explained in section 3.2., a link budget based on a CL averaged over three antenna orientationsprovidesasafetymarginwhich is4dBsuperiortoabudgetcalculatedwiththevaluemeasured inthebestorientation.
d➔ CL50%(d)=CL50%+10log(——)
2
d➔ CL95%(d)=CL95%+20log(——)
2
RadiatingCablesApplicationnote11/2021
4.1. Measurement procedures
TheproceduretomeasuretheCLisdefine byanIEC61196-4standard.Twoconfigu ationsarepermitted,i.e.:the“ground–levelmethod”andthe“free-spacemethod”.Thesetwoconfigu ationsoftengiveresultsthatmaybequitedifferent.That isnotsurprisingas it iswellknownthattheenvironmentaffectstheRCperformances.Asexplainedisthissection,theground–levelconfigu ationisclosertotheconditionsactuallyfoundintunnels.
Inaddition,thestandardallowstospecifyeitheraCLfora“singleorientation”(i.e.radial,orthogonalorparallel)orameanvaluecalculatedwithaspecifi formula.Thisissueisexaminedinsection4.2.
ThefactthattheIECstandardisnotveryrestrictingmaybeconfusing,especiallywhentheperformancespublishedinthemanufacturerdatasheetshavetobecompared.Someclarification areprovidedhereaftertoassisttheradioengineerinmakingthemostaccuratelinkbudgets.
4.1.1. Ground–level versus free-space methodThetwoconfigu ationsaredetailedintheannexBofthestandard(§B1.1and§B1.2)andareshowninFigures8and9respectively.
Intheground–levelmethod,theRCislaidat10to12cmaboveaconcreteground.Thecentreoftheantenna ispositionedverticallyat2mabovetheRC.Thefiel strength isrecordedwhenmovingtheantennaalongapathparalleltotheRC.
Inthefree-spacemethod,theRCishungtononmetallicpostsataheightof1.5to2m.Theantennacentreisat2mfromtheRCandatthesameheight.Thefiel strengthisrecordedwhenmovingtheantennaalongapathparalleltotheRC.
TheFigures8and9alsodefin thethreeantennaorientations,i.e.:➔ Radial: thedipoleisorientatedatrightanglewithrespecttotheRCandisinthesameplane;➔ Orthogonal: thedipoleisatrightanglewithrespecttotheplanecontainingtheRC;➔ Parallel: thedipoleisparalleltotheRC.
Figure8:RCandantennapositionswithground-levelmethod
Figure9:RCandantennapositionswithfree-spacemethod
12
4. COUPLINGLOSS
RadiatingCablesApplicationnote
11/2021
Figure10:Reflectio mechanism
withground-levelmethod
13
Ground-level and free-space methods sometimes give rather different CL results; to explain thesedifferences,coupledmodeandradiatedmodehavetobetreatedseparately.
Coupled mode cablesThedifferencebetweenCLswithground-levelandfree-spacemethodsmayberelativelyimportantandsometimesexceed10dB.Ingeneral,theground-levelmethodgiveslowerCLs;thisisnotsurprisingasthesurfaceclosetotheRCcontributestothecoupledmodegeneration.Inthefree-spaceconfigu ation,thegroundisat2mandistoofartoefficient ypromotethecoupledmode.
Itmustberemindedthattheradiatedmodecablesworkincoupledmodebelowthetransitionfrequency(whichdependsontheRCdesign).Consequently,theaboveremarksarealsoapplicabletotheseRCswhentheyareusedbelowtheirtransitionfrequency.
Radiated mode cablesFortheRCsworking inradiatedmode,CLdifferencesof2or3dBbetweenthetwoconfigu ationsareusualbutrarelyexceed6or7dB.Thisdifferencemaybeeitherpositiveornegative,dependingonRCdesignandfrequency.
TheCLdifferencesaremainlyduetotheeffectofthereflection onthegroundsurface.Indeed,intheground-levelconfigu ation,thereflection producedbytheconcretesurfacelocatedat10to12cmfromtheRChavearelativelyimportanteffect.Thereflectio mechanismisshowninFigure10whereonlyonesingleapertureAhasbeenconsideredforsimplicity.AtanypointPintheRCvicinity,thefiel strengthisthevectoradditionofthefiel radiatedbytheapertureA(hereaftertermed“directwave”)andtheonereflec edatthepointRbytheconcreteground.
Themagnitudeoftheresultingfiel willdependon:
➔ the magnitude of the reflected signal: thismagnitudedependson surfaceconductivity.Thereflectio coefficien mayrangebetween0(noreflection and1foraperfectlyconductivesurface.
➔ the phase difference between the direct and the reflected waves:thedirectandreflec edwavesdonottravelthesamedistance,henceaphasedifference.Itsvalue(indegrees)isgivenbytheexpression360°x(AR+RP–AP)/lwherelisthewavelengthintheair.
P
A
R
RC
Concreteground
RadiatingCablesApplicationnote11/2021
In addition there is a possible phase shift at the reflectio point R. This phase shift depends on thedirectionof theelectric fiel andon theelectricalpropertiesof theconcrete surface. In thecaseofa perfectly conductive surface, there is no phase shift for the component of the electric fiel whichisorientatedatrightanglewiththeconcreteground.Conversely,thecomponentoftheelectricfiel paralleltothegroundexperiencesa180°phaseshift.
Figure11showshowthereflection impacttheCL.InthisFigure,Ěd,Ěr,andĚ designatetheelectricfiel vectoratthepointPcorrespondingto,respectively,thedirectwave,thereflec edwaveandtheresultantfield TheleftpartofthisFigureshowsthecasewherethevectorscorrespondingtothedirectwaveĚdandreflec edwaveĚrarenearlyinphase.ThemagnitudeoftheirresultantĚbeinghigherthanĚd asthereflectio reinforcesthedirectwave,henceaCLdecrease.
Conversely,therightpartofFigure11showsthecasewherethevectorscorrespondingtothedirectwaveĚdandreflec edwaveĚrarenearlyinopposition.ThemagnitudeoftheirresultantĚislowerthanĚd,henceaCLincrease.
Although it has been assumed, in Figure 11, that the reflectio coefficien was lower than 1 (the ĚrvectorisshorterthanĚd),itisobviousthattheaboveconclusionsapplywhicheverthemagnitudeofthereflec edwave.
WiththewidebandRCs,itisnoteasytokeepthedirectandreflec edwavesinphase(ornearlyinphase)inallthefrequencybandsasthisparameterdependsonl.
Inthemostfavourablecase,i.e.whenthereisatotalreflectio (reflectio coefficien =1)inphasewiththedirectwave,theresultantĚ=2Ěd,hencea6dBCLdecrease.
Conversely, theworstcaseoccurswhere there isa total reflectio inoppositionwith thedirectwavebecausetheresultingfiel dropssharply,henceasevereCLincrease.Inpracticehowever,theresultingfiel doesnotcollapsecompletelyandtheCLincreaseshouldnotexceed20dB.
14
Figure11:Vectoradditionofthedirectandreflec edwaveswhentheyarenearlyinphase(ontheleft)andnearlyinopposition(ontheright).
Ed
^Ed
^
Er
^
Er
^
E
^
E
^
ĚdandĚrnearlyinphase ĚdandĚrnearlyinopposition
RadiatingCablesApplicationnote
11/2021
15
Comparedtoasituationwherethereisnoreflection theground-levelconfigu ationmayeitherproduceaCLdecreaseofmaximum6dBoranincreasethatshouldnotexceed20dB.
Inthefree-spaceconfigu ation,therearealsoreflection onthegroundsurfacebuttheireffectismuchless important as illustrated in Figure 12. Of course, if the RC is at 2 m above the concrete ground,thereflection canbeseenasproducedbyanelectrical image locatedatapproximately24.5moftheantenna. As the electromagnetic fiel decreases with the inverse of the distance, the magnitude ofthereflectio isabout0.44times (i.e.2m/4.5m)themagnitudeofdirectwavewhenthereflectio coefficien isequal to 1.Itresultsthat,ifthedirectandreflec edwavesareinphase,theCLdecrease3isabout3.2dB.Iftheyareinopposition,theCLincreasedoesnotexceed5dB.
Ifthereflectio coefficien islowerthan1,themagnitudeofthereflectio andtheimpactontheCLsisreducedaccordingly.
Itresultsthattheimpactofthereflection arelessimportantinthefree-spaceconfigu ationthanintheground-levelonewherethedirectandreflec edwavestravelnearlythesamedistance.Thisconclusionapplieswhicheverthevalueofthereflectio coefficient
4.1.2. Ground–level versus or-space methodInmostapplications,theRCishungatshortdistancefromasurface(ceilingorwall)producingreflection whichmayeither improveor impairtheCLs. It isobviousthattheground-levelmethod isclosertotheconditionsactuallymet inpractice.This is the reasonswhy theground-levelmethodseems themostsensibletoreferto.
4.1.3. Eupen RMC RangeTheEupenRMCrange isdesignedtoderivebenefi fromthereflectio phenomenon,at least inmostfrequencybandsallocatedtomobilecommunications.Thisisachievedbychoosingalaunchinganglethatminimisesthephasedifferencebetweendirectandreflec edwaves.
All the Eupen RC data sheets specify the CLs (50 and 95% probability) measured in the ground-levelconfigu ation. However, data sheets with the CLs measured in the free-space configu ation are alsoavailableformostEupenRCs.
2AccordingtoPythagora’stheorem,(4_+2_)1/2=4.5m320log(1+0.44)=3.2dBand20log(1-0,44)=5dB
Figure12:Reflectiomechanismwith
free-spacemethod R
RC
Concreteground
diplode
RadiatingCablesApplicationnote11/2021
Figure13:Half-wavedipoleresponse
4.2 Coupling loss and antenna orientations
4.2.1. CL definitions according to the standardTheIECstandardallowstospecifytheCLmeasuredeitherinasingleorientation(i.e.radial,orthogonalorparallel)orameanCLcalculatedwithaparticularformulagivenhereafter.Figures8and9showthethreeorientationsfortheground-levelandfree-spaceconfigu ationsrespectively.
MeasurementresultsindicatethattheCLdifferencebetweentheworst(highestCL)andthebest(lowestCL) orientationsmayexceed10andeven15dBinsomecases.TheexplanationofthiseffectisgiveninFigure13whereitisassumedthataverticallypolarisedelectromagneticfiel propagatesfromthelefttotherightasindicatedbythevectorv�.Thethreeconsideredantennaorientationsareandidentifie bythelettersa,bandc.
Iftheantennaarmsareorientatedhorizontallyandparalleltothedirectionofpropagation(lettera),theresponseshouldbetheoreticallynullbecausethemain lobeoftheradiationpattern ispointing intheverticalplane.
Iftheantennaarmsareorientatedhorizontallyandparalleltothedirectionofthemagneticfiel (letterb),theradiationpatternispointingtowardthesourceofthefiel buttheresponseshouldbetheoreticallynullbecausethearmsareperpendiculartotheelectricfield
Themaximumreceivedsignalisobtainedwiththeantennaarmsorientatedvertically(letterc).Indeed,theradiationpatternispointingtowardthesourceofthefiel andtheantennaarmsareparalleltotheelectricfield
The fact that the fiel produced by a RC is polarised explains the strong influenc of the antennaorientationontheCL.
As CL difference between orientations may exceed 10 and even 15 dB in the worst cases, correctunderstandingoftheimpactofthisparameterisrequiredforaccuratelinkbudgetcalculationsandwhentheperformancesofRCsfromdifferentmanufacturerhavetobecompared.
16
E
^
R
^
v
^
a b c
RadiatingCablesApplicationnote
11/2021
17
TheIECstandardalsodefine ameanCLcalculatedwiththefollowingformula:
CLmean=-10log[1_3
(10-CLr/10+10-CLo/10+10-CLp/10)]
This particular formula is different from the usual arithmetic and geometric averages. To understanditsphysicalmeaning, let’sconsideranRCfeedwithan inputpowerequalto1mW(0dBm).Theterm10–CLr/10intheaboveformulacorrespondstothepower(inmW)receivedbyadipoleantennaorientatedin the radial direction. Likewise, the terms 10 – CLo/10 and 10 – CLp/10 correspond to the power receivedbyadipoleantennaorientatedintheorthogonalandparalleldirectionsrespectively.
Consequently,theterm(10-CLr/10+10-CLo/10+10-CLp/10)/3isthesumofthepower(inmW)receivedwiththedipoleorientatedinthethreedifferentdirectionsdividedby3,i.e.thereceivedpoweraveragedonthethreeorientations.ItappearsthattheaboveformulagivesinfacttheCLwithrespecttothemeanvalueofthepowerreceivedinradial,verticalandorthogonalorientations.
Tounderstandtheimplicationsofthisdefinition let’sconsiderthesimplecasewheretheelectromagneticfiel isperfectlypolarisedinonedirection,forexampletheparallelone.ThisinvolvesthatonlyCLphasafini evaluewhileCLrandCLo=-∞.
As10-∞=0andas-10log[1_3
(10-CLp/10)]=-10log[1_3
]-10log[10-CLp/10]
Weobtainfinal y:
CLmean = 4.8 + CLp
Intheactualsituationshowever,thefiel isnearlyneverpurelypolarisedinonlyonedirection(i.e.thereisnodirectionforwhichthereceivedpowerisnull).Forinstance,ifCLr=60dB,CLo=70dBandCLp=70dB,weobtainCLmean=64dB.Othernumericalexamplesconfi mthattheCLmeanisgenerallyabout4dBhigherthanthelowestCL.Inconclusion,forpractice,thefollowingapproximationcanbemade:
CLmean =~ min (CLr, CLo, CLp) + 4
wherethe“min”symboldesignatestheminimumofthe3valuesinbrackets.
Although, thestandard imposes tospecify theantennaorientation, this information is lacking inmostmanufacturerdatasheets.Consequently,RCperformancescomparisonsaresometimesdifficul asthegivenCLcouldeitherbeameanvalueormeasuredinasingleunknownorientation.
RadiatingCablesApplicationnote11/2021
4.2.2. Antenna orientation and link budgetAsstatedabove,theCLinthe“worstorientation”maybe10to15dBhigherthaninthe“bestorientation”.Thisismainlyduetothefactthatthemeasuringantennaisahalf-wavedipolewhichfeaturesdirectivity.Indeed,theradiationpatternisthetypicaleightfigu ewithanullresponseinmedianplane.Consequently,theCLdependsonthedirectionofpropagationofthewaveradiatedbytheRCandontheorientationoftheelectricalfiel asshowninFigure13.
Inpracticehowever,theantennaorientationisnearlyneverperfectlyradialorparallelororthogonalwithrespecttotheRCbutratheracombinationofthesethreepossibilities. Indeed,themobileantenna isoftendowntiltedandisrarelyinanRCplane.
This remark also applies with handheld equipments. Moreover, their antenna is, generally, much lessdirectivethanadipole.Itmeansthattheirradiationpatternismore“isotropic”,resultinginadecreasedCLsensitivity to theantennaorientation.Consequently, theCLmean is recommended for linkbudgetcalculationinthecaseofcommunicationwithmobilephones.Inaddition,itmustalsoberemindedthatmobilephoneantennashavegainsubstantiallylowerthanthehalf-wavedipole.
Forallthesereasons,theEupenRMCdatasheetsspecifytheCL50%-meanandCL95%-mean.DetailedmeasurementreportswiththeCLsforthethreeorientationsareavailableonrequest.
18
RadiatingCablesApplicationnote
11/2021
19
5. RCPERFORMANCESOPTIMISATION
5.1. RC positioning
Themobileantennapositionandorientationareimportantparametersforperformancesoptimisationofradiocommunications inconfine spaces.Mobileantennamountedonthevehicleroofandhand-heldequipmentson-boardtrainarethemaincasesencounteredinpractice.Theyaredetailedhereafter.
5.1.1. Mobile antenna mounted on the vehicle roofFigure14showstwotypicalexampleswiththemobileantennainstalledonthevehicleroof(train,car,...).Itisgenerallyaquarterwavemonopoleorawhipverticallyorientedordowntilted.
Astheelectricfiel radiatedbytheRChasastrongradialcomponent,thebestcouplingisobtainedwiththeRChungfromthetunnelceilingandpreferablynearthecentrepositionoratleast1mawayfromthesidewallsasshowninFigure15.
The lowest CL and fiel strength fluctuation are obtained with the apertures located on the mobileside.Theaperture side ismarkedon theRC jacket.Figure12 shows theRCandaperturepositioningrecommendationsifthemobileantennaisinstalledonthevehicleroof.
Figure14:Mobileantennainstalled
onthevehicleroof
Figure15:RCpositionsifthemobileantennais
installedonthevehicleroof
min.1m aa
a
a=RecommendedzoneforRCposition Aperturesorientedtowardsvehicles
RadiatingCablesApplicationnote11/2021
Figure17:RecommendedRCpositionsforcommunicationwithpassengersonboardtrain
5.1.2. Hand-held mobile equipment on board trainItisobviousthattheorientationofahand-heldequipmentantennaisnearlyneververticalinnormaluseasshowninFigure16.
Theradiowavesenterscarriagesonlythroughthewindowswithapenetration losswhichdependsonglassmaterial(numberoflayers,metalcoating,...)andwindowsizes.
Inallcases,thebestcouplingisobtainedwiththeRChungalongawallasshowninFigure17(hand-heldontheRCsideandhand-heldontheoppositeside).ItisrecommendedtohangtheRCapproximatelyatthesameheightastheupperedgeofthecarriagewindowsasshowninthesefigu es.
Again,thelowestCLandfiel strengthfluctuation areobtainedwiththeapertureslocatedonmobileside.
20
Figure16:Hand-heldequipmentorientation
RadiatingCablesApplicationnote
11/2021
21
Wherethespecification imposeasecondRC,oneofthefollowingsolutionscanbeusedtomeetthereliabilityrequirementswithoutloosingthebenefi oflowfiel strengthfluctuations
➔ Feeding the 2 RCs with different carrier frequencies
The2RCsarefedwithdifferentcarrierfrequencysetsasshownintheseconddiagramofFigure18.ThusRC
1radiatesf
1onlyandRC
2radiatesf
3only.Thesameprincipleappliesfortheup-linkwithf
2andf
4.
➔ Use of only 1 RC at a time
OnlyRC1(mainlink)isactiveinnormaloperationconditionsasshowninthethirddiagramofFigure18.
RC2(sparelink)isactivatedincaseofRC
1linkfailure.
Figure18:Multi-cablesystem
RC1
RC2f1
RC1
RC1
GOOD
GOOD
RC2
RC2
f2
f3
f4
f1f3
f2f4
5.2. Multi-cable system
Insomecases,twoparallelRCsareusedinthesametubetoimprovethesystemreliability.Thediagramin the upper left corner of Figure 18 shows a configu ation where two nearby RCs are fed with thesameRF source.This solutionwillgive rise to large fiel strength fluctuation due to theconstructiveand destructive interferences between the signals radiated by the two RCs. Of course, the signal atfrequency f1 and f3 are simultaneously radiated by both RC1 and RC2. Hence, such a configu ationmust absolutely be avoided.
RadiatingCablesApplicationnote11/2021
RadiatingCables
11/2021Applicationnote
5.3.Resonant frequencies
Due to the fact that Radiating Mode Cables are based on a slot design with evenly spaced, repeating slots, it is a physical rule that this will lead to Resonances.
On Eupen’s EUCARAY® Radiating Cables, it results in clearly identified Resonant Frequencies. In most cases, the Resonant Frequencies occur in non-used Bands of the Radio Spectrum, and the even multiples are actually suppressed once the Cable has been installed. In addition, their magnitude generally decreases the higher the order.
It is good practice to verify that the Resonant Frequencies of a chosen EUCARAY® Radiating Cable do not appear in the required and used RF-Band. However, should a Resonant Frequency be found in the RF-Band to be used, it does not necessarily result in a Radiating Cable being unsuitable. On Eupen EUCARAY® Radiating Cable, Resonant Frequencies occupy only about 1 MHz of bandwidth, so that the transmission at frequencies located before and after a particular Resonant Frequency is not impacted. Further, the result at the said Resonant Frequency will typically be a slight increase in the Longitudinal Loss on the Downlink and a slightly raised VSWR in the Uplink. In both cases, the impact on the working of an RF System should be quite marginal. To compare, one should also bear in mind that a well-matched Antenna will have a typical VSWR of 1.5:1.
Resonant Frequencies versus Stop Bands
Resonant Frequencies on Eupen EUCARAY® Radiating Cables should not be compared to Stop Bands of other Manufacturers Cables. Such Stop Bands, or unauthorised Frequencies, usually occupy a much wider bandwidth, and the high level of reflected signal – with VSWR typically >> 10:1 – makes it impossible to use those Cables within the given frequency band.
The illustrations 1 & 2 below show the actual impact the Resonant Frequencies have on the VSWR and the Attenuation of one Eupen RMC-type Radiating Cable.(*)
The illustrations 3 & 4 show, at same scale, the impact of a Stop Band on a competitor cable.(*)
(*) Example with 1-5/8” size, broadband radiating cables.
22
EUPEN CABLE
YourreliablePartnerforRFCommunications
Asaleadingsupplieroftransmissionlinesandaccessoriestoglobalwirelesscommunicationsmarkets,EUPENhastheexperienceandresourcestoeffectivelyservicecustomersintoday’schallengingwirelesscommunicationsmarkets.
Atatimewhenwirelesscommunicationinconfinedareas,suchasunderground,streetandservicetunnels,isimportanttothenetworkoperators,EUPENhighqualityradiatingcablesprovethatEUPENCableisyourreliablePartnerforRFCommunications!
DownloadyourowncopyofthelatestRadiatingCablecatalogueat:www.radiating-cables.com
Tel. +32 87 597343 Fax. +32 87 597060
[email protected] [email protected]
23
RadiatingCablesApplicationnote11/2021
Malmedyer Str. 9 - B-4700 EUPEN - BELGIENTel.: +32(0)87.59.70.00 http://www.eupen.com Fax: +32(0)87.59.71.00 e-mail:[email protected]
ISO Certified Company